Chapter XV: Nervous System (2)

At a later period the blood-vessels of the pia mater form a rich plexus over the anterior part of the thin roof of the medulla, which becomes at the same time somewhat folded. The whole structure is known as the tela vasculosa, or choroid plexus of the fourth ventricle (fig. 250, _chd_ 4). The floor of the whole hind-brain becomes thickened, and there very soon appears on its outer surface a layer of non-medullated nerve-fibres, similar to those which first appear on the spinal cord. They are continuous with a similar layer of fibres on the floor of the mid-brain, where they constitute the crura cerebri. On the ventral floor of the medulla is a shallow continuation of the anterior fissure of the spinal cord.

In Elasmobranchii and many Teleostei the restiform tracts are well developed, and are anteriorly continued into the cerebellum, of which they form the peduncles. Near their junction with the cerebellum they form prominent bodies, which are regarded by Miklucho-Maclay as representing the true cerebellum of Elasmobranchii.

In Elasmobranchii a dorsal pair of ridges projects into the cavity of the fourth ventricle, corresponding apparently with the fasciculi teretes of the Mammalia.

In Mammalia there develop, subsequently to the longitudinal fibres already spoken of, first the olivary bodies of the ventral side of the medulla, and at a still later period the pyramids. The fasciculi teretes in the cavity of the fourth ventricle are developed shortly before the pyramids.

When the hind-brain becomes divided into two regions the roof of the anterior part does not become thinned out like that of the posterior, but on the contrary, becomes somewhat thickened and forms a band-like structure roofing over the anterior part of the fourth ventricle (fig. 247 and fig. 253, _cb_).

This is a rudiment of the cerebellum, and in all Craniate Vertebrates it at first presents this simple structure and insignificant size. In Cyclostomata, Amphibia and many Reptilia this condition is permanent. In Elasmobranchii, on the other hand, the cerebellum assumes in the course of development a greater and greater prominence (fig. 248, _cb_), and eventually overlaps both the optic lobes in front and the medulla behind. In the later embryonic stages it exhibits in surface-views the appearance of a median constriction, and the portion of the ventricle contained in it is prolonged into two lateral outgrowths.

Miklucho-Maclay, from his observations on the brains of adult Elasmobranchii, was led to regard what is here called the cerebellum as identical with the mid-brain, and the true mid-brain as part of the thalamencephalon. Miklucho-Maclay was no doubt misled by the large size of the cerebellum, but, as we have seen, this body does not begin to be conspicuous till late in embryonic life.

The mid-brain and thalamencephalon (according to the ordinary interpretations) have in the embryo of Elasmobranchs exactly the same relations as in the embryos of other Vertebrates; so that the embryological evidence appears to me to be conclusive against Miklucho-Maclay's view.

In Birds the cerebellum attains a very considerable development (fig. 250, _cbl_), consisting of a folded central lobe with an arbor vitæ, into which the fourth ventricle is prolonged. There are two small lateral lobes, apparently equivalent to the flocculi. Anteriorly the cerebellum is connected with the roof of the mid-brain by a delicate membrane, the velum medullæ anterius, or valve of Vieussens (fig. 250, _vma_). The pons Varolii of Mammalia is represented by a small number of transverse fibres on the floor of the hind-brain immediately below the cerebellum.

In Mammalia the cerebellum attains a still greater development. The median lobe or vermiform process is first developed. In the higher Mammalia the lateral parts forming the hemispheres of the cerebellum become formed as swellings at the sides at a considerably later period, and are hardly developed in the Monotremata and Marsupialia.

[FIG. 250. LONGITUDINAL SECTION THROUGH THE BRAIN OF A CHICK OF TEN
DAYS. (After Mihalkovics.)

_hms._ cerebral hemispheres; _alf._ olfactory lobe; _alf_{{1}}.
olfactory nerve; _ggt._ corpus striatum; _oma._ anterior commissure;
_chd{{3}}_. choroid plexus of the third ventricle; _pin._ pineal
gland; _cmp._ posterior commissure; _trm._ lamina terminalis; _chm._
optic chiasma; _inf._ infundibulum; _hph._ pituitary body; _bgm._
commissure of Sylvius (roof of iter a tertio ad quartum
ventriculum); _vma._ velum medullæ anterius (valve of Vieussens);
_cbl._ cerebellum; _chd{{4}}_. choroid plexus of the fourth
ventricle; _obl{{4}}_. roof of fourth ventricle; _obl._ medulla
oblongata; _pns._ commissural part of medulla; _inv._ sheath of
brain; _bls._ basilar artery; _crts._ internal carotid.]

The cerebellum is connected with the roof of the mid-brain in front and with the choroid plexus of the fourth ventricle behind by delicate membranous structures, known as the velum medullæ anterius (valve of Vieussens) and the velum medullæ posterius.

The pons Varolii is formed on the ventral side of the floor of the cerebellar region as a bundle of transverse fibres at about the same time as the olivary bodies.

The mid-brain. The changes undergone by the mid-brain are simpler than those of any other part of the brain. We have already seen that the mid-brain, on the appearance of the cranial flexure, forms _an unpaired vesicle_ with a vaulted roof and curved floor, at the front end of the long axis of the body (fig. 118, _MB_). It is at this period in most Vertebrates relatively much larger than in the adult; and it is only in the Teleostei that it more or less retains in the adult its embryonic proportions.

The cavity of the mid-brain, greatly reduced in size in the higher forms, is known as the iter a tertio ad quartum ventriculum, or aqueductus Sylvii.

The roof of the mid-brain is sharply constricted off from the divisions of the brain in front of and behind it, but these constrictions do not extend to the floor.

In some Vertebrates the region of the mid-brain is stated to undergo hardly any further development. In the Axolotl it remains according to Stieda[163] as a simple tube with nearly uniformly thick walls. In the majority of forms it undergoes, however, a more complicated development.

[163] "Ueb. d. Bau d. centralen Nervensystems d. Axolotl." _Zeit.
f. wiss. Zool._, Vol. xxv. 1875.

In Elasmobranchs the sides become thickened to form the optic lobes, which are soon separated by a median longitudinal groove. The floor becomes thickened to form the crura cerebri. The primitive simple median cavity becomes imperfectly divided into a median portion below, and two lateral diverticula in the optic lobes.

In Teleostei the changes, resulting in the formation of (1) a pair of longitudinal ridges projecting from the roof into the cavity of the iter, constituting the fornix of Gottsche, and (2) of the two swellings on the floor, forming the tori semicirculares, are more complicated, but have not been satisfactorily worked out. In Bombinator and the Anura generally the changes are of the same nature as those in Elasmobranchii, except that the prolongations of the ventricle into the optic lobes are still further constricted off from the median portion, which forms the true iter.

In Reptilia and Aves the development of the mid-brain takes place on the same type as in Elasmobranchii and the Anura. In Birds the optic lobes are pushed very much aside, and the roof of the iter is greatly thinned out. In Mammalia the sides of the mid-brain give rise to two pairs of prominences--the corpora quadrigemina--instead of the two optic lobes of other Vertebrata. The prominences, which do not contain prolongations of the iter, become first visible on the appearance of an oblique transverse furrow, while the anterior pair alone are separated by a longitudinal furrow. In the later stages of development the longitudinal furrow is continued so as to bisect the posterior pair.

The floor, which is bounded posteriorly by the pons Varolii, becomes the crura cerebri. The corpora geniculata interna also belong to this division of the brain.

Fore-brain. In its earliest condition the fore-brain forms a single vesicle without a trace of separate divisions, but very early it buds off the optic vesicles, whose history is described with that of the eye.

[FIG. 251. SECTION THROUGH THE FRONT PART OF THE HEAD OF A
LEPIDOSTEUS EMBRYO ON THE SEVENTH DAY AFTER IMPREGNATION.

_al._ alimentary tract; _fb._ thalamencephalon; _l._ lens of eye;
_op.v._ optic vesicle. The mesoblast is not represented.]

[FIG. 252. LONGITUDINAL SECTION THROUGH THE BRAIN OF A YOUNG
PRISTIURUS EMBRYO.

_cer._ commencement of cerebral hemisphere; _pn._ pineal gland;
_In._ infundibulum; _pt._ ingrowth of mouth to form the pituitary
body; _mb._ mid-brain; _cb._ cerebellum; _ch._ notochord; _al._
alimentary tract; _Iaa._ artery of mandibular arch.]

The optic vesicles become gradually constricted off from the fore-brain in a direction obliquely backwards and downwards. They remain, however, attached to it at the anterior extremity of the base of the fore-brain (fig. 251, _op.v._). While the above changes are taking place in the optic vesicles the anterior part of the fore-brain becomes prolonged, and at the same time somewhat dilated. At first there is no sharp boundary between the primitive fore-brain and its anterior prolongation, but there shortly appears a constriction which passes from above obliquely forwards and downwards. This constriction is shallow at first, but soon becomes much deeper, leaving however the cavities of the two divisions of the fore-brain united ventrally by a somewhat wide canal (fig. 252).

Of these two divisions the posterior becomes the thalamencephalon, while the anterior and larger division (_cer_) forms the rudiment of the cerebral hemispheres and olfactory lobes. For a considerable period this rudiment remains perfectly simple, and exhibits no signs, either externally or internally, of a longitudinal constriction dividing it into two lobes.

From the above description it may be concluded that the rudiment of the cerebral hemispheres is contained in the original fore-brain. In spite however of their great importance in all the Craniata, it is probable that the hemispheres were either not present as distinct structures, or only imperfectly separated from the thalamencephalon, in the primitive vertebrate stock.

The thalamencephalon. The thalamencephalon varies so slightly in structure throughout the Vertebrate series that a general description will suffice for all the types.

It forms at first a simple vesicle, the walls of which are of a nearly uniform thickness and formed of the usual spindle-shaped cells.

[FIG. 253. DIAGRAMMATIC VERTICAL SECTION THROUGH THE HEAD OF A LARVA
OF PETROMYZON.

The larva had been hatched three days, and was 4.8 mm. in length.
The optic and auditory vesicles are supposed to be seen through the
tissues.

_c.h._ cerebral hemisphere; _th._ optic thalamus; _in._
infundibulum; _pn._ pineal gland; _mb._ mid-brain; _cb._ cerebellum;
_md._ medulla oblongata; _au.v._ auditory vesicle; _op._ optic
vesicle; _ol._ olfactory pit; _m._ mouth; _br.c._ branchial pouches;
_th._ thyroid involution; _v.ao._ ventral aorta; _ht._ ventricle of
heart; _ch._ notochord.]

The cavity it contains is known as the third ventricle. Anteriorly it opens widely into the cerebral rudiment, and posteriorly into the ventricle of the mid-brain. The opening into the cerebral rudiment becomes the foramen of Munro.

For convenience of description I shall divide it into three regions, viz. (1) the floor, (2) the sides, and (3) the roof.

The floor becomes divided into two parts, an anterior part, giving origin to the optic nerves, in which is formed the optic chiasma; and a posterior part, which becomes produced into an at first inconspicuous prominence--the rudiment of the infundibulum (fig. 252, _In_). This comes in contact with an involution from the mouth, which gives rise to the pituitary body (fig. 252, _pt_), the development of which will be dealt with separately.

In the later stages of development the infundibulum becomes gradually prolonged, and forms an elongated diverticulum of the third ventricle, the apex of which is in contact with the pituitary body (figs. 252, 254, _in_, and figs. 250 and 255, _inf_).

Along the sides of the infundibulum run the commissural fibres connecting the floor of the mid-brain with the cerebrum.

In its later stages the infundibular region presents considerable variations in the different vertebrate types. In Fishes it generally remains very large, and permanently forms a marked diverticulum of the floor of the thalamencephalon. In Elasmobranchii the distal end becomes divided into three lobes--a median and two lateral. The lateral lobes appear to become the sacci vasculosi of the adult.

In Teleostei peculiar bodies known as the lobi inferiores (hypoaria) make their appearance at the sides of the infundibulum. They appear to correspond in position with the tuber cinereum of Mammalia[164]. In Birds, Reptiles, and Amphibia the lower part of the embryonic infundibulum becomes atrophied and reduced to a mere finger-like process--the processus infundibuli.

[164] For the relations of these bodies, _vide_ L. Stieda, "Stud.
üb. d. centrale Nervensystem d. Knochenfische." _Zeit. f. wiss.
Zool._ Vol. XVIII. 1868.

In Mammalia the posterior part of the primitive infundibulum becomes the corpus albicans, which is double in Man and the higher Apes; the ventral part of the posterior wall forms the tuber cinereum. Laterally, at the junction of the optic thalami and infundibulum, there are placed the fibres of the crura cerebri, which are probably derived from the walls of the infundibulum. A special process grows out from the base of the infundibulum, which undergoes peculiar changes, and becomes intimately united with the pituitary body; in which connection it will be more fully described.

[FIG. 254. LONGITUDINAL SECTION THROUGH THE BRAIN OF SCYLLIUM
CANICULA AT AN ADVANCED STAGE OF DEVELOPMENT.

_cer._ cerebral hemisphere; _pn._ pineal gland; _op. th._ optic
thalamus, connected with its fellow by a commissure (the middle
commissure). In front of it is seen a fold of the roof of the
fore-brain, which is the choroid plexus of the third ventricle;
_op._ optic chiasma; _pt._ pituitary body; _in._ infundibulum; _cb._
cerebellum; _au.v._ passage leading from the auditory vesicle to the
exterior; _mel._ medulla oblongata; _c.in._ internal carotid
artery.]

The sides of the thalamencephalon become very early thickened to form the optic thalami, which constitute the most important section of the thalamencephalon. They are separated, in Mammalia at all events, on their inner aspect from the infundibular region by a somewhat S-shaped groove, known as the sulcus of Munro, which ends in the foramen of Munro. They also become in Mammalia secondarily united by a transverse commissure, the grey or middle commissure, which passes across the cavity of the third ventricle. This commissure is probably homologous with, and derived from, a commissural band in the roof of the thalamencephalon, placed immediately in front of the pineal gland which is well developed in Elasmobranchii (fig. 254).

The roof undergoes more complicated changes. It becomes divided, on the appearance of the pineal gland as a small papilliform outgrowth (the development of which is dealt with separately), into two regions--a longer anterior in front of the pineal gland and a shorter posterior. The anterior region becomes at an early period excessively thin, and at a later period, when the roof of the thalamencephalon is shortened by the approach of the cerebral hemispheres to the mid-brain, it becomes (_vide_ figs. 250 and 255, _chd_ 3, and 254) considerably folded, while at the same time a vascular plexus is formed in the pia mater above it. On the accomplishment of these changes it is known as the tela choroidea of the third ventricle.

In the roof of the third ventricle behind the pineal gland there appear in Elasmobranchii, the Sauropsida and Mammalia transverse commissural fibres, forming a structure known as the posterior commissure, which connects together the two optic thalami.

The most remarkable organ in the roof of the thalamencephalon is the pineal gland, which is developed in most Vertebrates as a simple papilliform outgrowth of the roof, and is at first composed of cells similar to those of the other parts of the central nervous system (figs. 250, 252, 254 and 255, _pn_ or _pin_). In the lower Vertebrata it is directed forwards, but in Mammalia, and to some extent in Aves, it is directed backwards.

In Amphibia it is described by Götte (No. 296) as being a product of the point where the roof of the brain remains latest attached to the external skin.

The figure which Götte gives to prove this does not appear to me fully to bear out his conclusion; which if true is very important. Although I directed my attention specially to this point, I could find no indication in Elasmobranchii of a process similar to that described by Götte, and his observations have not as yet been confirmed for other Vertebrates. Götte compares the pineal gland to the long-persisting pore which leads into the cavity of the brain in the embryo of Amphioxus, and we might add the Ascidians, and, should his facts be confirmed, the conclusion he draws from them would appear to be well founded.

The later stages in the development of the pineal gland in different Vertebrates have not in all cases been fully worked out[165].

[165] For a full account of this subject _vide_ Ehlers (No. 337).

[FIG. 255. LONGITUDINAL VERTICAL SECTION THROUGH THE ANTERIOR PART
OF THE BRAIN OF AN EMBRYO RABBIT OF FOUR CENTIMETRES. (After
Mihalkovics.)

The section passes through the median line so that the cerebral
hemispheres are not cut; their position is however indicated in
outline.

_spt._ septum lucidum formed by the coalescence of the inner walls
of part of the cerebral hemispheres; _cna._ anterior commissure;
_frx._ vertical pillars of the fornix; _cal._ genu of corpus
callosum; _trm._ lamina terminalis; _hms._ cerebral hemispheres;
_olf._ olfactory lobes; _acl._ artery of corpus callosum; _fmr._
position of foramen of Munro; _chd{{3}}._ choroid plexus of third
ventricle; _pin._ pineal gland; _cmp._ posterior commissure; _bgm._
lamina uniting the lobes of the mid-brain; _chm._ optic chiasma;
_hph._ pituitary body; _inf._ infundibulum; _pns_. pons Varolii;
_pde._ cerebral peduncles; _agd._ iter.]

In Elasmobranchii the pineal gland becomes in time very long, and extends far forwards over the roof of the cerebral hemispheres (fig. 254 _pn_). Its distal extremity dilates somewhat, and in the adult the whole organ forms (Ehlers, No. 337) an elongated tube, enlarged at its free extremity, and opening at its base into the brain. The enlarged extremity may either be lodged in a cavity in the cartilage of the cranium (Acanthias), or be placed outside the cranium (Raja).

In Petromyzon its form is very different. It arises (fig. 253 _pn_) as a sack-like diverticulum of the thalamencephalon extending at first both backwards and forwards. In the Ammocoete the walls of this sack are deeply infolded.

The embryonic form of the pineal gland in Amphibia is very much like that which remains permanent in Elasmobranchii; the stalk connecting the enlarged terminal portion with the brain soon however becomes solid and very thin except at its proximal extremity. The enlarged portion also becomes solid, and is placed in the adult externally to the skull, where it forms a mass originally described by Stieda as the cerebral gland.

In Birds the primitive outgrowth to form the pineal gland becomes, according to Mihalkovics, deeply indented by vascular connective tissue ingrowths, so that it assumes a dendritic structure (fig. 250 _pin_).

The proximal extremity attached to the roof of the thalamencephalon forms a special section, known as the infra-pineal process. The central lumen of the free part of the gland finally atrophies, but the branches still remain hollow. The infra-pineal process becomes reduced to a narrow stalk, connecting the branched portion of the body with the brain. The branched terminal portion and the stalk obviously correspond with the vesicle and distal part of the stalk of the types already described. In Mammalia the development of the pineal gland is, according to Mihalkovics, generally similar to that of Birds. The original outgrowth becomes branched, but the follicles or lobes to which the branching gives rise eventually become solid (fig. 255 _pin_). An infra-pineal process is developed comparatively late, and is not sharply separated from the roof of the brain.

No satisfactory suggestions have yet been offered as to the nature of the pineal gland, unless the view of Götte be regarded as such. It appears to possess in all forms an epithelial structure, but, except at the base of the stalk (infra-pineal process) in Mammalia, in the wall of which there are nerve-fibres, no nervous structures are present in it in the adult state.

The pituitary body. Although the pituitary body is not properly a nervous structure, yet from its intimate connection with the brain it will be convenient to describe its development here. The pituitary body is in fact an organ derived from the epiblast of the stomodæum. This fact has been demonstrated for Mammalia, Aves, Amphibia and Elasmobranchii, and may be accepted as holding good for all the Craniata[166]. The epiblast in the angle formed by the cranial flexure becomes involuted to form the cavity of the mouth. This cavity is bordered on its posterior surface by the front wall of the alimentary tract, and on its anterior by the base of the fore-brain. Its uppermost end does not at first become markedly constricted off from the remainder, but is nevertheless the rudiment of the pituitary body.

[166] Scott states that in the larva of Petromyzon the pituitary
body is derived from the walls of the nasal pit; _Quart. J. of
Micr. Science_, Vol. XXI. p. 750. I have not myself completely
followed its development in Petromyzon, but I have observed a
slight diverticulum of the stomodæum which I believe gives origin
to it. Fuller details are in any case required before we can
admit so great a divergence from the normal development as is
indicated by Scott's statements.

Fig. 256 represents a transverse section through the head of an Elasmobranch embryo, in which, owing to the cranial flexure, the fore part of the head is cut longitudinally and horizontally, and the section passes through both the fore-brain (_fb_) and the hind-brain. Close to the base of the fore-brain are seen the mouth (_m_), and the pituitary involution from this (_pt_). In contact with the pituitary involution is the blind anterior termination of the throat (_al_) which a little way back opens to the exterior by the first visceral cleft (1 _v.c._). This figure alone suffices to demonstrate the correctness of the above account of the pituitary body; but its truth is still further confirmed by fig. 252; in which the mouth involution (_pt_) is in contact with, but still separated from, the front end of the alimentary tract. Very shortly after the septum between the mouth and throat becomes pierced, and the two are placed in communication, the pituitary involution becomes very partially constricted off from the mouth involution, though still in direct communication with it. In later stages the pituitary involution becomes longer and is dilated terminally; while the passage connecting it with the mouth becomes narrower and narrower, and is finally reduced to a solid cord, which in its turn disappears.

Before the connection between the pituitary vesicle and the mouth is obliterated the cartilaginous cranium becomes developed, and it may then be seen that the infundibulum projects through the pituitary space to come into close juxtaposition with the pituitary body.

After the pituitary vesicle has lost its connection with the mouth it lies just in front of the infundibulum (figs. 250 and 255 _hph_ and fig. 254 _pt_); and soon becomes surrounded by vascular mesoblast, which grows in and divides it into a number of branching tubes. In many forms the cavity of the vesicle completely disappears, and the branches become for the most part solid [Cyclostomata and some Mammalia (the rabbit), Elasmobranchii, Teleostei and Amphibia]. In Reptilia, Aves and most Mammalia the lumen of the organ is more or less retained (W. Müller, No. 344).

[FIG. 256. TRANSVERSE SECTION THROUGH THE FRONT PART OF THE HEAD OF
A YOUNG PRISTIURUS EMBRYO.

The section, owing to the cranial flexure, cuts both the fore- and
the hind-brain. It shews the premandibular and mandibular head
cavities _1pp_ and _2pp_, etc. The section is moreover somewhat
oblique from side to side.

_fb._ fore-brain; _l._ lens of eye; _m._ mouth; _pt._ upper end of
mouth, forming pituitary involution; _1ao._ mandibular aortic arch;
_1pp._ and _2pp._ first and second head cavities; _1vc._ first
visceral cleft; _V._ fifth nerve; _aun._ auditory nerve; _VII._
seventh nerve; _aa._ roots of dorsal aorta; _acv._ anterior cardinal
vein; _ch._ notochord.]

Although in the majority of the Vertebrata there is a close connection between the pituitary body and the infundibulum, there is no actual fusion between the two. In Mammalia the case is different. The part of the infundibulum which lies at the hinder end of the pituitary body is at first a simple finger-like process of the brain (fig. 255 _inf_), but its end becomes swollen, and the lumen in this part becomes obliterated. Its cells, originally similar to those of the other parts of the nervous system and even (Kölliker) containing differentiated nerve-fibres, partly atrophy, and partly assume an indifferent form, while at the same time there grow in amongst them numerous vascular and connective-tissue elements. The process of the infundibulum thus metamorphosed becomes inseparably connected with the true pituitary body, of which it is usually described as the posterior lobe. The part of the infundibulum which undergoes this change is very probably homologous with the saccus vasculosus of Fishes.

The true nature of the pituitary body has not yet been made out. It is clearly a rudimentary organ in existing craniate Vertebrates, and its development indicates that when functional it was probably a sense organ opening into the mouth, its blind end reaching to the base of the brain. No similar organ has as yet been found in Amphioxus, but it seems possible perhaps to identify it with the peculiar ciliated sack placed at the opening of the pharynx in the Tunicata, the development of which was described at p. 18. If the suggestion is correct, the division of the body into lobes in existing Vertebrata must be regarded as a step towards a retrogressive metamorphosis.

Another possible view is to regard the pituitary body as a glandular structure which originally opened into the mouth in the lower Chordata, but which has in all existing forms ceased to be functional. The intimate relation of the organ to the brain appears to me opposed to this view of its nature, while on the other hand its permanent structure is more easily explained on this view than on that previously stated. In the Ascidians a glandular organ has been described by Lacaze Duthiers[167] in juxtaposition to the ciliated sack, and it is possible that this organ as well as the ciliated sack may be related to the pituitary body. In view of this possibility further investigations ought to be carried out in order to determine whether the whole pituitary body is derived from the oral involution, or whether there may not be a nervous part and a glandular part of the organ.

[167] "Les Ascidies simples des Côtes de France." _Archives de
Biologie expér. et générale_, Vol. III. 1874, p. 329.

The Cerebral Hemispheres. It will be convenient to treat separately the development of the cerebral hemispheres proper, and that of the olfactory lobes.

Although the cerebral hemispheres vary more than any other part of the brain, they are nevertheless developed from the unpaired cerebral rudiment in a nearly similar manner throughout the series of Vertebrata.

In the cerebral rudiment two parts may be distinguished, viz. the floor and the roof. The former gives rise to the ganglia at the base of the hemispheres--corpora striata, etc.--the latter to the hemispheres proper.

[FIG. 257. DIAGRAMMATIC LONGITUDINAL HORIZONTAL SECTION THROUGH THE
FORE-BRAIN.

_3.v._ third ventricle; _lv._ lateral ventricle; _lt._ lamina
terminalis; _ce._ cerebral hemisphere; _op.th._ optic thalamus.]

The first change which takes place consists in the roof growing out into two lobes, between which a shallow median constriction makes its appearance (fig. 257). The two lobes thus formed are the rudiments of the two hemispheres. The cavity of each of them opens by a widish aperture into the vestibule at the base of the cerebral rudiment, which again opens directly into the cavity of the third ventricle (_3 v_). The Y-shaped aperture thus formed, which leads from the cerebral hemispheres into the third ventricle, is the foramen of Munro. The cavity (_lv_) in each of the rudimentary hemispheres is a lateral ventricle. The part of the cerebrum which lies between the two hemispheres, and passes forwards from the roof of the third ventricle round the end of the brain to the optic chiasma, is the rudiment of the lamina terminalis (figs. 257 _lt_ and 255 _trm_). Up to this point the development of the cerebrum is similar in all Vertebrata, but in some forms it practically does not proceed much further.

In Elasmobranchii, although the cerebrum reaches a considerable size (fig. 254 _cer_), and grows some way backwards over the thalamencephalon, yet it is not in many forms divided into two distinct lobes, but its paired nature is only marked by a shallow constriction on the surface. The lamina terminalis in the later stages of development grows backwards as a thick median septum which completely separates the two lateral ventricles[168] (fig. 263).

[168] A comparison of the mode of development of this septum with
that of the septum lucidum with its contained commissures in
Mammalia clearly shews that the two structures are not
homologous, and that Miklucho-Maclay is in error in attempting to
treat them as being so.

There are, it may be mentioned, considerable variations in the structure of the cerebrum in Elasmobranchii into which it is not however within the scope of this work to enter.

In the Teleostei the vesicles of the cerebral hemispheres appear at first to have a wide lumen, but it subsequently becomes almost or quite obliterated, and the cerebral rudiment forms a small bilobed nearly solid body. In Petromyzon (fig. 253 _ch_) the cerebral rudiment is at first an unpaired anterior vesicle, which subsequently becomes bilobed in the normal manner. The walls of the hemispheres become much thickened, but the lateral ventricles persist.

In all the higher Vertebrates the division of the cerebral rudiment into two distinct hemispheres is quite complete, and with the deepening of the furrow between the two hemispheres the lamina terminalis is carried backwards till it forms a thin layer bounding the third ventricle anteriorly, while the lateral ventricles open directly into the third ventricle.

In Amphibians the two hemispheres become united together immediately in front of the lamina terminalis by commissural fibres, forming the anterior commissure. They also send out anteriorly two solid prolongations, usually spoken of as the olfactory lobes, which subsequently fuse together.

In all Reptilia and Aves there is formed an anterior commissure, and in the higher members of the group, especially Aves (fig. 250), the hemispheres may obtain a considerable development. Their outer walls are much thickened, while their inner walls become very thin; and a well-developed ganglionic mass, equivalent to the corpus striatum, is formed at their base.

The cerebral hemispheres undergo in Mammalia the most complicated development. The primitive unpaired cerebral rudiment becomes, as in lower Vertebrates, bilobed, and at the same time divided by the ingrowth of a septum of connective tissue into two distinct hemispheres (figs. 260 and 261 _f_ and 258 1). From this septum is formed the falx cerebri and other parts.

The hemispheres contain at first very large cavities, communicating by a wide foramen of Munro with the third ventricle (fig. 260). They grow rapidly in size, and extend, _especially backwards_, and gradually cover the thalamencephalon and the mid-brain (fig. 258 1, _f_). The foramen of Munro becomes very much narrowed and reduced to a mere slit.

[FIG. 258. BRAIN OF A THREE MONTHS' HUMAN EMBRYO: NATURAL SIZE.
(From Kölliker.)

1. From above with the dorsal part of hemispheres and mid-brain
removed; 2. From below. _f._ anterior part of cut wall of the
hemisphere; _f´._ cornu ammonis; _tho._ optic thalamus; _cst._
corpus striatum; _to._ optic tract; _cm._ corpora mammillaria; _p._
pons Varolii.]

The walls are originally nearly uniformly thick, but the floor becomes thickened on each side, and gives rise to the corpus striatum (figs. 260 and 261 _st_). The corpus striatum projects upwards into each lateral ventricle, giving to it a somewhat semilunar form, the two horns of which constitute the permanent anterior and descending cornua of the lateral ventricles (fig. 262 _st_).

[FIG. 259. TRANSVERSE SECTION THROUGH THE BRAIN OF A RABBIT OF FIVE
CENTIMETRES. (After Mihalkovics.)

The section passes through nearly the posterior border of the septum
lucidum, immediately in front of the foramen of Munro.

_hms._ cerebral hemispheres; _cal._ corpus callosum; _amm._ cornu
ammonis (hippocampus major); _cms._ superior commissure of the
cornua ammonis; _spt._ septum lucidum; _frx 2._ vertical fibres of
the fornix; _cma._ anterior commissure; _trm._ lamina terminalis;
_str._ corpus striatum; _ltf._ nucleus lenticularis of corpus
striatum; _vtr 1._ lateral ventricle; _vtr 3._ third ventricle;
_ipl._ slit between cerebral hemispheres.]

With the further growth of the hemisphere the corpus striatum loses its primitive relations to the descending cornu. The reduction in size of the foramen of Munro above mentioned is, to a large extent, caused by the growth of the corpora striata.

[FIG. 260. TRANSVERSE SECTION THROUGH THE BRAIN OF A SHEEP'S EMBRYO
OF 2.7 CM. IN LENGTH. (From Kölliker.)

The section passes through the level of the foramen of Munro.

_st._ corpus striatum; _m._ foramen of Munro; _t._ third ventricle;
_pl._ choroid plexus of lateral ventricle; _f._ falx cerebri; _th._
anterior part of optic thalamus; _ch._ optic chiasma; _o._ optic
nerve; _c._ fibres of the cerebral peduncles; _h._ cornu ammonis;
_p._ pharynx; _sa._ presphenoid bone; _a._ orbitosphenoid bone; _s._
points to part of the roof of the brain at the junction between the
roof of the third ventricle and the lamina terminalis; _l._ lateral
ventricle.]

The corpora striata are united at their posterior border with the optic thalami. In the later stages of development the area of contact between these two pairs of ganglia increases to an immense extent (fig. 261), and the boundary between them becomes somewhat obscure, so that the sharp distinction which exists in the embryo between the thalamencephalon and cerebral hemispheres becomes lost. This change is usually (Mihalkovics, Kölliker) attributed to a fusion between the corpora striata and optic thalami, but it has recently been attributed by Schwalbe (No. 349), with more probability, to a growth of the original surface of contact, and an accompanying change in the relations of the parts.

The outer wall of the hemispheres gradually thickens, while the inner wall becomes thinner. In the latter, two curved folds, projecting towards the interior of the lateral ventricle, become formed. These folds extend from the foramen of Munro along nearly the whole of what afterwards becomes the descending cornu of the lateral ventricle.

The upper fold becomes the hippocampus major (cornu ammonis) (figs. 259 _amm_, 260 and 261 _h_, and 262 _am_). When the rudiment of the descending cornu has become transformed into a simple process of the lateral ventricle the hippocampus major forms a prominence upon its floor.

[FIG. 261. TRANSVERSE SECTION THROUGH THE BRAIN OF A SHEEP'S EMBRYO
OF 2.7 CM. IN LENGTH. (From Kölliker.)

The section is taken a short distance behind the section represented
in fig. 260, and passes through the posterior part of the
hemispheres and the third ventricle.

_st._ corpus striatum; _th._ optic thalamus; _to._ optic tract; _t._
third ventricle; _d._ roof of third ventricle; _c._ fibres of
cerebral peduncles; _c´._ divergence of these fibres into the walls
of the hemispheres; _e._ lateral ventricle with choroid plexus _pl_;
_h._ cornu ammonis; _f._ primitive falx; _am._ alisphenoid; _a._
orbitosphenoid; _sa._ presphenoid; _p._ pharynx; _mk._ Meckel's
cartilage.]

The wall of the lower fold becomes very thin, and a vascular plexus, derived from the connective-tissue septum between the hemispheres, and similar to that of the roof of the third ventricle, is formed outside it. It constitutes a fold projecting far into the cavity of the lateral ventricle, and together with the vascular connective tissue in it gives rise to the choroid plexus of the lateral ventricle (figs. 260 and 261 _pl_).

It is clear from the above description that a marginal fissure leading into the cavity of the lateral ventricle does not exist in the sense often implied in works on human anatomy, in that the epithelium covering the choroid plexus, which forms the true wall of the brain, is a continuous membrane. The _epithelium_ of the choroid plexus of the lateral ventricle is quite independent of that of the choroid plexus of the third ventricle, though at the foramen of Munro the roof of the third ventricle is of course continuous with the inner wall of the lateral ventricle (fig. 260 _s_). The _vascular elements_ of the two plexuses form however a continuous structure.

The most characteristic parts of the Mammalian cerebrum are the commissures connecting the two hemispheres. These commissures are (1) the anterior commissure, (2) the fornix, and (3) the corpus callosum, the two latter being peculiar to Mammalia.

By the fusion of the inner walls of the hemispheres in front of the lamina terminalis a solid septum is formed, known as the septum lucidum, continuous behind with the lamina terminalis, and below with the corpora striata (figs. 255 and 259 _spt_). It is by a series of differentiations within this septum that the above commissures originate. In Man there is a closed cavity left in the septum known as the fifth ventricle, which has however no communication with the true ventricles of the brain.

In the septum lucidum there become first formed, below, the transverse fibres of the anterior commissure (fig. 255 and fig. 259 _cma_), and in the upper part the vertical fibres of the fornix (fig. 255 and fig. 259 _frx 2_). The vertical fibres meet above the foramen of Munro, and thence diverge backwards, as the posterior pillars, to lose themselves in the cornu ammonis (fig. 259 _amm_). Ventrally they are continued, as the descending or anterior pillars of the fornix, into the corpus albicans, and thence into the optic thalami.

The corpus callosum is not formed till after the anterior commissure and fornix. It arises in the upper part of the region (septum lucidum) formed by the fusion of the lateral walls of the hemispheres (figs. 255 and 259 _cal_), and at first only its curved anterior portion--the genu or rostrum--is developed. This portion is alone found in Monotremes and Marsupials. The posterior portion, which is present in all the Monodelphia, is gradually formed as the hemispheres are prolonged further backwards.

[FIG. 262. LATERAL VIEW OF THE BRAIN OF A CALF EMBRYO OF 5 CM.
(After Mihalkovics.)

The outer wall of the hemisphere is removed, so as to give a view of
the interior of the left lateral ventricle.

_hs._ cut wall of hemisphere; _st._ corpus striatum; _am._
hippocampus major (cornu ammonis); _d._ choroid plexus of lateral
ventricle; _fm._ foramen of Munro; _op._ optic tract; _in._
infundibulum; _mb._ mid-brain; _ch._ cerebellum; _IV.V._ roof of
fourth ventricle; _ps._ pons Varolii, close to which is the fifth
nerve with Gasserian ganglion.]

Primitively the Mammalian cerebrum, like that of the lower Vertebrata, is quite smooth. In many of the Mammalia, _Monotremata_, _Insectivora, etc._, this condition is nearly retained through life, while in the majority of Mammalia a more or less complicated system of fissures is developed on the surface. The most important, and first formed, of these is the Sylvian fissure. It arises at the time when the hemispheres, owing to their growth in front of and behind the corpora striata, have assumed a somewhat bean-shaped form. At the root of the hemispheres--the hilus of the bean--there is formed a shallow depression, which constitutes the first trace of the Sylvian fissure. The part of the brain lying in this fissure is known as the island of Reil.

The olfactory lobes. The olfactory lobes, or rhinencephala, are secondary outgrowths of the cerebral hemispheres, and contain prolongations of the lateral ventricles, but may however be solid in the adult state. According to Marshall they develop in Birds and Elasmobranchs and presumably other forms later than the olfactory nerves, so that the olfactory region of the hemispheres is indicated before the appearance of the olfactory lobes.

In most Vertebrates the olfactory lobes arise at a fairly early stage of development from the under and anterior part of the hemispheres (fig. 250 _olf_). In Elasmobranchs they arise, not from the base, but from the lateral parts of the brain (fig. 263), and become subsequently divided into a bulbous portion and a stalk. They vary considerably in their structure in the adult.

[FIG. 263. SECTION THROUGH THE BRAIN AND OLFACTORY ORGAN OF AN
EMBRYO OF SCYLLIUM. (Modified from figures by Marshall and myself.)

_ch._ cerebral hemispheres; _ol.v._ olfactory vesicle; _olf._
olfactory pit; _Sch._ Schneiderian folds; _I._ olfactory nerve. The
reference line has been accidentally taken through the nerve to the
brain; _pn._ anterior prolongation of pineal gland.]

In Amphibia the solid anterior prolongations of the cerebral hemispheres already spoken of are usually regarded as the olfactory lobes, but according to Götte, whose view appears to me well founded, small papillæ, situated at the base of these prolongations, from which olfactory nerves spring, and which contain a process of the lateral ventricle, should properly be regarded as the olfactory lobes. These papillæ arise prior to the solid anterior prolongations of the hemispheres.

In Birds the olfactory lobes are small. In the chick they arise (Marshall) on the seventh day of incubation.

_General conclusions as to the Central Nervous System._

It has been shewn above that both the brain and spinal cord are primitively composed of a uniform wall of epithelial cells, and that the first differentiation results in the formation of an external layer of white matter, a middle layer of grey matter (ganglion cells), and an inner epithelial layer. This primitive histological arrangement, which in many parts of the brain at any rate, is only to be observed in the early developmental stages, has a simple phylogenetic explanation.

As has been already explained in an earlier part of this chapter the central nervous system was originally a differentiated part of the superficial epidermis.

This differentiation (as may be concluded from the character of the nervous system in the Coelenterata and Echinodermata) consisted in the conversion of the inner ends of the epithelial cells into nerve-fibres; that is to say, that the first differentiation resulted in the formation of a layer of white matter on the inner side of the epidermis. The next stage was the separation of a deeper layer of the epidermis as a layer of ganglion cells from the superficial epithelial layer, _i.e._ the formation of a middle layer of ganglion cells and an outer epithelial layer. Thus, phylogenetically, the same three layers as those which first make their appearance in the ontogeny of the vertebrate nervous system became successively differentiated, and in both cases they are clearly placed in the same positions, because the central canal of the vertebrate nervous system, as formed by an involution, is at the true outer surface, and the external part of the cord is at the true inner surface.

It is probable that a very sharp distinction between the white and grey matter is a feature acquired in the higher Vertebrata, since in Amphioxus there is no such sharp separation; though the nerve-fibres are mainly situated externally and the nerve-cells internally.

As already stated in Chapter XII. the primitive division of the nervous axis was probably not into brain and spinal cord, but into (1) a fore-brain, representing the ganglion of the præoral lobe, and (2) the posterior part of the nervous axis, consisting of the mid- and hind-brains and the spinal cord. This view of the division of the central nervous system fits in fairly satisfactorily with the facts of development. The fore-brain is, histologically, more distinct from the posterior part of the nervous system than the posterior parts are from each other; the front end of the notochord forms the boundary between these two parts of the central nervous system (_vide_ fig. 253), ending as it does at the front termination of the floor of the mid-brain, and finally, the nerves of the fore-brain have a different character to those of the mid- and hind-brain.

This primitive division of the central nervous system is lost in all the true Vertebrata, and in its place there is a secondary division--corresponding with the secondary vertebrate head--into a brain and spinal cord. The brain, as it is established in these forms, is again divided into a fore-brain, a mid-brain and a hind-brain. The fore-brain is, as we have already seen, the original ganglion of the præoral lobe. The mid-brain appears to be the lobe, or ganglion, of the third pair of nerves (first pair of segmental nerves), while the hind-brain is a more complex structure, each section of which (perhaps indicated by the constrictions which often appear at an early stage of development) giving rise to a pair of segmental nerves is, roughly speaking, homologous with the whole mid-brain.

The type of differentiation of each of the primitively simple vesicles forming the fore-, the mid- and the hind-brains is very uniform throughout the Vertebrate series, but it is highly instructive to notice the great variations in the relative importance of the parts of the brain in the different types. This is especially striking in the case of the fore-brain, where the cerebral hemispheres, which on embryological grounds we may conclude to have been hardly differentiated as distinct parts of the fore-brain in the most primitive types now extinct, gradually become more and more prominent, till in the highest Mammalia they constitute a more important section of the brain than the whole of the remaining parts put together.

The little that is known with reference to the significance of the more or less corresponding outgrowths of the floor and roof of the thalamencephalon, constituting the infundibulum and pineal gland, has already been mentioned in connection with the development of these parts.

BIBLIOGRAPHY.

(332) C. J. Carus. _Versuch einer Darstellung d. Nervensystems, etc._ Leipzig, 1814.

(333) J. L. Clark. "Researches on the development of the spinal cord in Man, Mammalia and Birds." _Phil. Trans._, 1862.

(334) E. Dursy. "Beiträge zur Entwicklungsgeschichte des Hirnanhanges." _Centralblatt f. d. med. Wissenschaften_, 1868. Nr. 8.

(335) E. Dursy. _Zur Entwicklungsgeschichte des Kopfes des Menschen und der höheren Wirbelthiere._ Tübingen, 1869.

(336) A. Ecker. "Zur Entwicklungsgeschichte der Furchen und Windungen der Grosshirn-Hemisphären im Foetus des Menschen." _Archiv f. Anthropologie_, v. Ecker und Lindenschmidt. Vol. III. 1868.

(337) E. Ehlers. "Die Epiphyse am Gehirn d. Plagiostomen." _Zeit. f. wiss. Zool_. Vol. XXX., suppl. 1878.

(338) P. Flechsig. _Die Leitungsbahnen im Gehirn und Rückenmark des Menschen. Auf Grund entwicklungsgeschichtlicher Untersuchungen._ Leipzig, 1876.

(339) V. Hensen. "Zur Entwicklung des Nervensystems." _Virchow's Archiv_, Bd. XXX. 1864.

(340) L. Löwe. "Beiträge z. Anat. u. z. Entwick. d. Nervensystems d. Säugethiere u. d. Menschen." Berlin, 1880.

(341) L. Löwe. "Beiträge z. vergleich. Morphogenesis d. centralen Nervensystems d. Wirbelthiere." _Mittheil. a. d. embryol. Instit. Wien_, Vol. II. 1880.

(342) A. M. Marshall. "The Morphology of the Vertebrate Olfactory organ." _Quart. J. of Micr. Science_, Vol. XIX. 1879.

(343) V. v. Mihalkovics. _Entwicklungsgeschichte d. Gehirns_. Leipzig, 1877.

(344) W. Müller. "Ueber Entwicklung und Bau der Hypophysis und des Processus infundibuli cerebri." _Jenaische Zeitschrift_. Bd. VI. 1871.

(345) H. Rahl-Rückhard. "Die gegenseitigen Verhältnisse d. Chorda, Hypophysis etc. bei Haifischembryonen, nebst Bemerkungen üb. d. Deutung d. einzelnen Theile d. Fischgehirns." _Morphol. Jahrbuch_, Vol. VI. 1880.

(346) H. Rathke. "Ueber die Entstehung der glandula pituitaria." _Müller's Archiv f. Anat. und Physiol._, Bd. V. 1838.

(347) C. B. Reichert. _Der Bau des menschlichen Gehirns._ Leipzig, 1859 u. 1861.

(348) F. Schmidt. "Beiträge zur Entwicklungsgeschichte des Gehirns." _Zeitschrift f. wiss. Zoologie_, 1862. Bd. XI.

(349) G. Schwalbe. "Beitrag z. Entwick. d. Zwischenhirns." _Sitz. d. Jenaischen Gesell. f. Med. u. Naturwiss._ Jan. 23, 1880.

(350) Fried. Tiedemann. _Anatomie und Bildungsgeschichte des Gehirns im Foetus des Menschen._ Nürnberg, 1816.

THE DEVELOPMENT OF THE CRANIAL AND SPINAL NERVES[169].

[169] Remak derived the posterior ganglia from the tissue of the
mesoblastic somites, and following in Remak's steps most authors
believed the peripheral nervous system to have a mesoblastic
origin. This view, which had however been rejected on theoretical
grounds by Hensen and others, was finally attacked on the ground
of observation by His (No. 297). His (No. 352, p. 458) found that
in the Fowl "the spinal ganglia of the head and trunk arose from
a small band of matter which is placed between the medullary
plate and epiblast, and the material of which he called the
'intermediate cord'." He further states that: "Before the closure
of the medullary tube this band forms a special groove--the
'intermediate groove'--placed close to the border of the
medullary plate. As the closure of the medullary plate into a
tube is completed, the earlier intermediate groove becomes a
compact cord. In the head of the embryo a longitudinal ridge
arises in this way, which separates the suture of the brain from
that of the epiblast. In the parts of the neck and in the
remaining region of the neck the intermediate cord does not lie
over the line of junction of the medullary tube, but laterally
from this and forms a ridge, triangular in section, with a slight
indrawing." This intermediate ridge gives rise to four ganglia in
the head, viz. the g. trigemini, g. acousticum, g.
glossopharyngei, and g. vagi, and in the trunk to the spinal
ganglia. In both cases it unites first with the spinal cord.

I have given in the above account, as far as possible, a literal
translation of His' own words, because the reader will thus be
enabled fairly to appreciate his meaning.

Subsequently to His' memoir (No. 297) I gave an account of some
researches of my own on this subject (No. 351), stating the whole
of the nerves to be formed as cellular outgrowths of the spinal
cord. I failed fully to appreciate that some of the stages I
spoke of had been already accurately described by His, though
interpreted by him very differently. Marshall, and afterwards
Kölliker, arrived at results in the main similar to my own, and
Hensen, independently of and nearly simultaneously with myself,
published briefly some observations on the nerves of Mammals in
harmony with my results.

His has since worked over the subject again (No. 352), and has
reaffirmed as a result of his work his original statements. I
cannot, however, accept his interpretations on the subject, and
must refer the reader who is anxious to study them more fully, to
His' own paper.

All the nerves are outgrowths of the central nervous system, but the differences in development between the cranial and spinal nerves are sufficiently great to make it convenient to treat them separately.

Spinal nerves. The posterior roots of the spinal nerves, as well as certain of the cranial nerves, arise in the same manner, and from the same structure, and are formed considerably before the anterior roots. Elasmobranch fishes may be taken as the type to illustrate the mode of formation of the spinal nerves.

The whole of the nerves in question arise as outgrowths of a median ridge of cells, which makes its appearance on the dorsal side of the spinal cord (fig. 264 A, _pr_). This ridge has been called by Marshall the neural crest. At each point, where a pair of nerves will be formed, two pear-shaped outgrowths project from it, one on each side; and apply themselves closely to the walls of the spinal cord (fig. 264 B, _pr_). These outgrowths are the rudiments of the posterior nerves. While still remaining attached to the dorsal summit of the neural cord they grow to a considerable size (fig. 264 B, _pr_).

[FIG. 264 A. TRANSVERSE SECTION THROUGH A PRISTIURUS EMBRYO SHEWING
THE PROLIFERATION OF CELLS TO FORM THE NEURAL CREST.

_pr._ neural crest; _nc._ neural canal; _ch._ notochord; _ao._
aorta.]

[FIG. 264 B. TRANSVERSE SECTION THROUGH THE TRUNK OF AN EMBRYO
SLIGHTLY OLDER THAN FIG. 28 E.

_nc._ neural canal; _pr._ posterior root of spinal nerve; _x._
subnotochordal rod; _ao._ aorta; _sc._ somatic mesoblast; _sp._
splanchnic mesoblast; _mp._ muscle-plate; _mp´._ portion of
muscle-plate converted into muscle; _Vv._ portion of the vertebral
plate which will give rise to the vertebral bodies; _al._ alimentary
tract.]

[FIG. 265. VERTICAL LONGITUDINAL SECTION THROUGH PART OF THE TRUNK
OF A YOUNG SCYLLIUM EMBRYO.

_com._ commissure uniting the dorsal ends of the posterior
nerve-roots; _pr._ ganglia of posterior roots; _ar._ anterior roots;
_st._ segmental tubes; _sd._ segmental duct; _g.e._ epithelium
lining the body cavity in the region of the future germinal ridge.]

The attachment to the dorsal summit is not permanent, but before describing the further fate of the nerve-rudiments it is necessary to say a few words as to the neural crest. At the period when the nerves have begun to shift their attachment to the spinal cord, there makes its appearance, in Elasmobranchii, a longitudinal commissure connecting the dorsal ends of all the spinal nerves (figs. 265, 266 _com_), as well as those of the vagus and glossopharyngeal nerves. This commissure has as yet only been found in a complete form in Elasmobranchii; but it is nevertheless to be regarded as a very important morphological structure.

[FIG. 266. SPINAL NERVES OF SCYLLIUM IN LONGITUDINAL SECTION TO SHEW
THE COMMISSURE CONNECTING THEM.

A. Section through a series of nerves.
B. Highly magnified view of the dorsal part of a single nerve, and
of the commissure connected with it.

_com._ commissure; _sp.g._ ganglion of posterior root; _ar._
anterior root.]

It is probable, though the point has not yet been definitely made out, that this commissure is derived from the neural crest, which appears therefore to separate into two cords, one connected with each set of dorsal roots.

[FIG. 267. SECTION THROUGH THE DORSAL PART OF THE TRUNK OF A TORPEDO
EMBRYO.